Reliability test porting: stringy stuff....

Reliability test porting: stringy stuff.

No functional change. This gets things compiling on linux, but it still needs some work before it can be run on linux.

BUG=19085
Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=23272

[Evan Stade, 2009-08-12 19:57:11]

adding more reviewers

[tony, 2009-08-12 20:04:14]

Adding mark so he can review the file_path changes.  He can probably comment on
the logging changes too.

[Paweł Hajdan Jr., 2009-08-12 20:07:52]

A bunch of nits...

http://codereview.chromium.org/164352/diff/1015/29
File chrome/test/reliability/page_load_test.cc (right):

http://codereview.chromium.org/164352/diff/1015/29#newcode167
Line 167: #endif
Add // defined(WIN) here.

http://codereview.chromium.org/164352/diff/1015/29#newcode214
Line 214: #endif
// defined (OS_WIN)

http://codereview.chromium.org/164352/diff/1015/29#newcode343
Line 343: if (timeout_ms == -1)
Use base::kNoTimeout here.

http://codereview.chromium.org/164352/diff/1015/29#newcode466
Line 466: #endif
Add // defined(OS_WIN)

http://codereview.chromium.org/164352/diff/1015/29#newcode512
Line 512: ASSERT_TRUE(file_util::Delete(crash_dump_file_path, false));
consider using DieFileDie from test_file_util.h

It might have impact on flakiness.

http://codereview.chromium.org/164352/diff/1015/29#newcode549
Line 549: #endif
// defined(OS_WIN)

http://codereview.chromium.org/164352/diff/1015/29#newcode635
Line 635: #endif
// defined(OS_WIN)

[Mark Mentovai, 2009-08-12 20:12:41]

I didn't read page_load_test as carefully as the rest, which looks good to me
with these changes.

http://codereview.chromium.org/164352/diff/1015/32
File chrome/common/logging_chrome.cc (right):

http://codereview.chromium.org/164352/diff/1015/32#newcode192
Line 192: #if defined(OS_WIN)
This doesn't need to be an #ifdef any longer.

http://codereview.chromium.org/164352/diff/1015/29
File chrome/test/reliability/page_load_test.cc (right):

http://codereview.chromium.org/164352/diff/1015/29#newcode89
Line 89: std::string server_url = "http://urllist.com";
Existing, but we shouldn't have any globals of std::string - this runs code, the
constructor, before main().

[Evan Stade, 2009-08-12 20:53:41]

http://codereview.chromium.org/164352/diff/1015/32
File chrome/common/logging_chrome.cc (right):

http://codereview.chromium.org/164352/diff/1015/32#newcode192
Line 192: #if defined(OS_WIN)
On 2009/08/12 20:12:41, Mark Mentovai wrote:
> This doesn't need to be an #ifdef any longer.

Done.

http://codereview.chromium.org/164352/diff/1015/29
File chrome/test/reliability/page_load_test.cc (right):

http://codereview.chromium.org/164352/diff/1015/29#newcode466
Line 466: #endif
On 2009/08/12 20:07:52, Paweł Hajdan Jr. wrote:
> Add // defined(OS_WIN)

Done.

http://codereview.chromium.org/164352/diff/1015/29#newcode512
Line 512: ASSERT_TRUE(file_util::Delete(crash_dump_file_path, false));
On 2009/08/12 20:07:52, Paweł Hajdan Jr. wrote:
> consider using DieFileDie from test_file_util.h
> 
> It might have impact on flakiness.

Done.

http://codereview.chromium.org/164352/diff/1015/29#newcode549
Line 549: #endif
On 2009/08/12 20:07:52, Paweł Hajdan Jr. wrote:
> // defined(OS_WIN)

Done. (and the rest)

[Evan Stade, 2009-08-12 20:54:14]

http://codereview.chromium.org/164352/diff/1015/29
File chrome/test/reliability/page_load_test.cc (right):

http://codereview.chromium.org/164352/diff/1015/29#newcode89
Line 89: std::string server_url = "http://urllist.com";
On 2009/08/12 20:12:41, Mark Mentovai wrote:
> Existing, but we shouldn't have any globals of std::string - this runs code,
the
> constructor, before main().

Done.

[Evan Stade, 2009-08-12 23:55:11]

(lg from Pawel in chat)